Age-related macular degeneration (AMD) is the leading cause of visual loss in the elderly. Despite knowledge of the cell types involved, therapeutic intervention has been limited and there is currently no treatment for nonexudative AMD. While RPE cell transplantation into the subretinal space of patients offered a promising therapeutic approach, outcomes to date have been limited due to: 1) transplantation in late stage disease, 2) the invasive route of administration, and 3) incomplete differentiation status of the transplanted cells. To address these limitations, we have made a number of exciting discoveries: 1) forced expression of the RPE65 gene allows mouse hematopoietic stem cells (mHSC), when injected back into the circulation, to home to the retina and renew the RPE monolayer in both acute and chronic mouse models of RPE loss and re-establish visual function; 2) the circadian pattern to endogenous HSC release impacts reconstitution following bone marrow transplantation; 3) microglial activation in AMD will require modulation to ensure efficient RPE regeneration by HSCs; 4) a highly effective non-viral protein delivery machinery (T3SS) is able to deliver target proteins into host HSCs and promote their differentiation; 5) human hematopoietic progenitor cells (hHPCs), when programmed with RPE65, express RPE cell markers. Based on these observations we hypothesize that: Successful therapeutic utilization of human or murine HSC requires their programming prior to injection into the systemic circulation, their injection at the time of optimal engraftment potential and preconditioning of the retina by either suppression of resident microglia activation and/or restoring the balance of peripheral pro-inflammatory and homeostatic monocytes. The hypothesis is addressed in three Aims. In Aim 1 we will determine the dependence of recruitment and incorporation of programmed mHSC into the injured RPE in the SOD2 KD mouse model upon the time of day of injection and the age of the donor HSC as well as the age of the recipient. Aim 2 will investigate the importance of manipulating the retinal environment, by controlling either the activation state of the resident microglia or the influx of peripheral monocytes on the efficiency of systemically administered programmed mHSC to repair the RPE layer in the SOD2 KD model. Aim 3 will translate our mouse findings into hHPCs. We will express RPE65, in human CD133+, CD 34- , CD38- cells to differentiate these cells toward RPE cells and allow RPE regeneration in SCID mice undergoing the SOD2 KD model. Co-injection of mesenchymal stem cells will be utilized to reduce activation of resident microglia. Our approach overcomes many of the current limitations of human stem cell therapies for AMD.